Continued support of the Great Lakes New England (GLNE) Clinical Epidemiology and Validation (CEV) of the Early Detection Research Network (EDRN) via the U01 mechanism is requested. The GLNE-CEV consists of the a collaboration of the following Institutions and their respective consortium partners: University of Michigan Medical Center- Cancer Center, MD Anderson Cancer Center, Dartmouth Medical School-Norris Cotton Cancer Center, the Dana Farber Cancer Institute, and St. Michael Hospital of Toronto (University of Toronto). In the next grant period, the GLNE-CEV proposes to 1. Continue development and validation of biomarker panels for the early detection and risk assessment of epithelial cancers of the GI tract;and 2. Continue assembly of a high quality, prospectively collected, protocol driven, carefully documented biorepository to support EDRN collaborative biomarker discovery and validation research. Aim #1 will be addressed through the development and implementation of biomarker validation protocols for four different GI tract sites: colorectal adenocarcinoma, pancreatic adenocarcinoma, esophageal adenocarcinoma, and hepatocellular carcinoma. Working groups for each of these sites have been established. For colorectal adenocarcinoma detection, the GLNE-CEV proposes development of a stool based panel of genetic biomarkers supplemented by serum based proteomic and mucin derived biomarkers. For pancreatic adenocarcinoma detection, the GLNE-CEV proposes a panel of serum based proteomics discovered biomarkers followed by endoscopic ultrasound guided fine needle biopsy of cells with PCR detection of high risk genes discovered by genomic profiling. For low esophagus early detection, the GLNE-CEV proposes a panel of serum-based methylation biomarkers coupled with tissue based ploidy, cyclin Dl, and a genetic amplicon panel. For hepatocellular detection, an Associate Member affiliated with the GLNE proposes a validation trial of descarboxyprothrombin supplemented by proteomic profiles and proteomics discovered biomarkers. Aim #2 will be addressed by a multi-institutional network system that develops detailed standard operating procedures and data elements. The data are managed via a Good Computing Practice compliant, Web-fronted informatics network that includes a sample tracking system and a secure protocol server. Biosamples are managed in a tracked, managed, secure biorepository. The GLNE-CEV consortium, while focused upon the validation of biomarkers for colorectal cancer provides extensive, provides diverse human populations from which to validate early detection biomarkers for high priority cancers.